@danielunderwood In the interval [2000, 10000] will you probably have a good fit.

Yeah you could fit it in a variety of ways if you have the data, but I was looking at the method for fitting over batches that come in over time (think you get 5-minute batches every 5 minutes). I could probably make it a bit nicer by fixing the fits to be continuous and adding some of the previous data with small weights. Might try that out when I feel less lazy

Morning

Last night India beat Australia in WC, scoring 352/5 50.0 Overs and Australia scoring 316/10 50 overs

I had a suspicion, after putting naive GR and quantum mechanics together:

Secret's secret suspects

Claim: We can actually change the past, without producing a parallel timeline

Consider a naive quantisation of the metric $g$ such that $g$ can be written as a superposition of metrics:

In particular, $g$ is not globally hyperbolic and contain a CTC between events A and B as seen by observer O

When the observer is at $A$ in his worldline, he received a journal article which he deduced it is from the future, hence he can model the spacetime in his future light cone using the above expression

where each of the $g(\phi)$ corresponds to different possible position of the event $B$ where the journal article originated from

Now let's suppose some time in the future, O did something that result in the journal article to never be sent back in the first place. Then the event of the journal article will not necessary get undone because it could possibly be sent back later in the future.

And at the limit where there is no more possible delay before the journal article must be sent back, will then interfering it becomes impossible because the event A is already determined by a measurement (O's mailbox for example)

But consider the case where it is O himself that send the article back, then because of the superpositions of events B, it is possible for O to never fulfill the role of closing the loop because the event B will not be gone until right before O's death

which is also a probabilistic event

And thus, you have this weird scenario that in order to maintain self consistency, O's life can be prolonged indefinitely since his death before the article is sent back will violate self consistency and thus he cannot die by effectively stretching the CTC to forever

wahhhhh

...A second thought on the above however, seemed to suggest it is too deliberate. It seems more likely the below situation will happen instead, meaning you can only delay a predestinated future, only up to right before the event will then become self inconsistent, or worse, the events A and B are already entangled, thus upon receiving the journal article, the measurement had already taken place on $g$, projecting to one of the $g(\phi)$ s where event A and B is fixed and well defined,

and O will have no way to know about it

I actually suspect if time travel is possible and under quantum gravity situations, CTCs might actually behave like a pair of entangled events stretching across timelike distances

Thus put it bluntly, the moment you have interacted with something from the future that is uncaused, you will have no choice but to close the loop at some determined but unpredictable time in the future

except maybe, you construct some crazy $g'$ with some kind of gravity technology such that the state $g$ changes, thus breaking the correlation between events A and B due to the measurement at A (which is really a joint measurement because A and B are entangled due to the CTC structure)

and thereby, freeing you from the CTC at the cost of a topology change in spacetime (which is not known to be allowed in classical GR)

If this is the case, then time travel may be even more interesting in the quantum gravity regime because then you can scramble causal diagrams of events so badly such that no two diagrams between two timelike separated regions of spacetime are connected, and thus as long you expend energy to change the topology of spacetime, you can change the past however you like without causing any paradoxes

But if that is the case, does time have meaning anymore globally speaking...?

Anyway, this is just speculation based on the existing theories plus some naive generalisation, I don't think quantum gravity, being in such a strong interacting regime, will be a linear theory, thus superposition may not hold in general

morning

Hello

it's dusk now.

cricket-world-cup-2019 July 14th is the final

@skullpatrol the existence of people who care about cricket continues to befuddle me

Fortunately they were all confined to an island

So as to spare the world from their nonsense

@EmilioPisanty As opposed to

:P

@ACuriousMind wow! must've been good party to keep you away for that long?

@ACuriousMind the man wants me to preorder 2077

Damn the man

@PM2Ring LOL looks like he had enough forums.xkcd.com/viewtopic.php?f=18&t=126714#p4460649

@RyanUnger Preorder is the root of all evil

@ACuriousMind I haven’t preordered in years

Welcome to 𝕰𝖓𝖌𝖑𝖆𝖓𝖉

same over here

it was like 30° for two days and now it's just thunderstorms and rain

Weather while I was cycling back from the shops today - have a guess! :-)

@JohnRennie Meteors?

@Slereah speaking of which apparently someone's just identified a large meteor crater in Germany.

Ah, now I take the trouble to Google it the crater was identified 50 years ago. Oh well, the blog post gave the impression it had only just been identified.

@ACuriousMind damn that cold

Those are of course Celsius degrees :P

Question, does the near-unity value of (Plank Constant)/(3*pi^2*c^3*[electron mass]^2) make sense in a context I haven't played with yet? (Although I don't think it is actually within error bars of unity, unless I am miscalculating uncertainties, there might be a constant relationship I haven't considered yet)

(I ask because I was creating a ratio between reduced Compton wavelength and mass in crackpot gravity stuff, and got approximately 1.5pi instead of the E-11 value I was kind of expecting, and am trying to figure out if the set of things I plugged in have a relationship I didn't anticipate, apart from the one between mass and Compton wavelength, which failed entirely to clear things up for me)

@JohnRennie you saw meteor rock?

@Adirian unity in what units? Your numerator has units of energy * time and your denominator has units of energy^2 * time/length. So that ratio would be length/(energy * time)

@Adirian the numerical value of a constant that has dimensions is not especially meaningful because it depends on the units i.e. it depends on the definition of the metre, second, etc. So for a dimensionful constant to have a value close to one is generally not that interesting. More interesting are the values of dimensionless constants because they do not depend on the units we choose.

For instance, the fact that the proton rest energy is approximately one GeV is a useful fact but hardly a profound one

@JohnRennie I just finished The Colour of Magic and I'm looking at which discworld audiobook to listen to next. I'm not sure if I really want to follow on with The Light Fantastic right now, and was thinking about either Guards! Guards! or Mort. IIRC you're a pretty big discworld fan, any suggestions?

I'll take that as a "No, you aren't ignoring a relationship". Cool. I was worried I was canceling out something without realizing it.

It's c^2, I just realized, not c^3, though. The final expected units are s^2/(kg*m)

And the mass isn't squared. Christ. I had it right in my math but not my notation above the math

Or maybe not in my math. Nevermind all that, I need to figure out what on Earth I did

h/(m c^2) would have units of time. (In fact that might even be the Planck time? I forget.)

Ah, I already did substitutions, is what I did. Divided the reduced Compton wavelength by (c^2*Me), then started trying to figure out why that gave me the value it did and started substituting.

@JMac Mort is excellent, and so is Wyrd Sisters.

I wasn't so keen on Guards! Guards! though friends of mine love it so that's probably just my particular taste.

Ok, no, h/mc^2 isn’t the Planck time. (It contains the electron mass which isn’t fundamental so ofc it’s not)

Planck units are constructed using c, G and $\hbar$

Also Boltzmann’s constant and Coulomb’s constant.

@JohnRennie Well that tips the scale for me. I was leaning a bit towards Mort anyways, because it's part of the "Death" books, and Death seems like a pretty cool character (I hear he gets better in other books, but I found him pretty entertaining anyways in Colour of magic)

@JMac Mort is an excellent book. It's one of his very early ones.

I think with the later books Pratchett got tempted into trying to make points with his books, and they lack the easy spontaneity of the early books.

One thing I noticed is that a lot of people (online) seem to dislike the part in colour of magic where suddenly he's a doctor on a plane stopping a terrorist attack. I don't know why, but I found that scene completely hilarious how we suddenly go from dragon riding to a terrorist plane plot

@JohnRennie Funny, I like the later ones more precisely because they don't have that spontaneous "slapstick" feel to them the early ones do, but instead tell more cohesive stories

@ACuriousMind if we all liked the same stuff there wouldn't be anything to argue about :-)

I'm usually all for cohesive stories, but I do find this has been a nice change of pace from the usual plot-heavy fantasy I've been reading

I'm currently reading Dracul by Dacre Stoker (he's related to Bram Stoker in some way I can't remember) and it's rubbish, but highly entertaining rubbish.

hello creatures of God

@JohnRennie Sure, but in this case you're simply objectively wrong ;)

hii @JohnRennie

@JohnRennie That sounds like a total cash-grab. Is he basically Dune-ifying it?

@JMac no, from what I've read so far it is an interesting idea and not simply derivative of the original Dracula novel. But I doubt it hurts to be related to Bram Stoker if you're trying to sell books about vampires.

It seems like something someone with the last name Stoker might want to avoid if they didn't want to appear to be cashing in on name recognition. I guess it's a bit of shaky territory if he really wanted to write a Dracula book; but it definitely comes across on first glance as writing Dracula works because of his last name.

Anyhow, I'm going to pour myself a beer and get back to the book. I have to finish it before the end of the week as a SF reading group I go to are discussing it this Saturday.

So, folks

good news

Knotting fractional-order knots with the polarization state of light. E Pisanty et al. Nature Photonics, doi:10.1038/s41566-019-0450-2 (2019). Courtesy eprint here.

Multicolored light twists in new knotted ways (ICFO press release)

I assume you're one of the few people here whose avatar will ever be featured as a figure in a scientific paper.

Also, congratulations. :-)

@HDE226868 thx =)

@HDE226868 not the first time this happens. You just have to keep your profile pic ahead of the formal publications ;-)

Ha, I'll have to keep that in mind for the future.

If memory serves, this is the third figure from a paper that I use.

@EmilioPisanty I understand some of those words. Congrats though. I imagine nature photonics is probably like one of the bigger journals in your field?

@JMac close to as big as it gets :-)

@EmilioPisanty That's pretty awesome.

@HDE226868 there's someone from your school here, is it you?

@RyanUnger Where?

(So probably no, but I might know them.)

@HDE226868 Princeton

@RyanUnger Not me, but I know some recent grads from my school who are going there. One's going into chem, and the other is a year into . . . sorta like cogsci but with a bunch of other things thrown in.

@RyanUnger Is the person's first name Ben(jamin) or Mayank?

@HDE226868 oh no I meant at a summer school

@RyanUnger Oh, interesting. Well, still not me.

Can you upvote/downvote your own answers?

@Avantgarde no.